Journal of Applied Microbiology 2000, 88, 1001ÿ1008
The effect of temperature and selective agents on the growth of
Yersinia enterocolitica serotype O:3 in pure culture
C.M. Logue1, J.J. Sheridan2, D.A. McDowell3, I.S. Blair3 and T. Hegarty4
1
Department of Veterinary and Microbiological Sciences, North Dakota State University, Fargo, ND, USA, 2The National
Food Centre, Teagasc, Dublin, Ireland, 3Food Studies Research Unit, The University of Ulster, Jordanstown,
Newtownabbey, Northern Ireland and 4Statistics Department, Teagasc, Dublin, Ireland
81/12/99: received 1 December 1999, revised 9 February 2000 and accepted 11 February 2000
This study
examined the individual and combined effects of the selective agents normally present in
Yersinia-selective agar (i.e. cefsulodin, irgasan and novobiocin) on the growth kinetics of
plasmid-bearing (P‡) and plasmid-cured (P±) Yersinia enterocolitica serotype O:3 at 25 and
37 C. Growth studies were carried out in pure culture, and the data obtained were
subjected to linear regression analysis to determine lag phase duration(s) and growth rates of
the examined strains. In general, the presence of selective agents increased the duration of
the lag phase at 37 C, with longer lag phases noted in all cases in which two or more
selective agents were present. Growth rates in CIN broth base (CIN NA) and CIN NA plus
commercial supplement (SR 109) (CIN) were faster at 37 than 25 C, but in some cultures
of incomplete CIN NA broth with less than three supplements added, growth tended to be
faster at 25 than 37 C. Generally, plasmid-bearing strains grew slower than plasmid-cured
strains in most media at 37 C due to virulence plasmid expression retarding growth. In
some instances at 37 C, it was observed that the growth rates of both plasmid-bearing and
plasmid-cured strains were comparable, indicating the in¯uence of added selective agent/s
negating any effects associated with virulence plasmid expression. The effects of selective
agents, incubation temperature and virulence plasmid carriage on the growth kinetics of Y.
enterocolitica are discussed.
C . M . L O G U E , J . J . S H E R I D A N , D . A . M C D O W E L L , I . S . B L A I R A N D T . H E G A R T Y . 2000.
INTRODUCTION
Yersinia enterocolitica is a signi®cant foodborne pathogen
causing gastroenteritis, and a range of severe sequalea in
humans (Schiemann 1989). It is therefore important, to
have an effective and rapid means for its isolation and or
identi®cation from suspect foods, environmental and clinical samples. Currently, the most common procedure
involves selective plating on Yersinia-selective agar (CIN)
(Schiemann 1979, 1982).
Schiemann (1980) investigated the growth of Yersinia in
the presence of selective agents during studies to develop
Yersinia-selective agar. Other workers have also examined
the growth of Yersinia on a range of media (Fukushima
and Gomyoda 1986; Toora et al. 1994). There is, however,
Correspondence to: C.M. Logue, Department of Veterinary and
Microbiological Sciences, 130 A Van Es Hall, North Dakota State University,
Fargo, ND 58105, USA (e-mail: [email protected]).
= 2000 The Society for Applied Microbiology
little reliable data regarding the growth kinetics of Yersinia
in such media, and the effect of the selective agents, i.e.
cefsulodin, irgasan and novobiocin on this pathogen.
A number of studies (McDermott et al. 1993; Goverde
et al. 1994) have shown that the choice of incubation temperature affects the growth kinetics of plasmid-bearing and
plasmid-cured strains of Escherichia coli and Y. enterocolitica. In general, virulence plasmid expression in Yersinia
occurs at temperatures > 30 C, this expression is usually
observed in phenotypic characteristics such as calcium
dependency, autoagglutination and production of outer
membrane proteins (Laird and Cavanaugh 1980; Cornelis
et al. 1987; Bhaduri et al. 1990). Goverde et al. (1994)
reported that virulence plasmid expression at temperatures
greater than 30 C retards the growth rate of plasmid-bearing Y. enterocolitica strains as plasmid replication places an
additional metabolic burden on growing cells.
A previous investigation (Sheridan et al. 1998) reported
that the growth rate of plasmid-bearing Y. enterocolitica
1002 C . M . L O G U E E T A L .
strains were signi®cantly faster at 37 C than plasmid-cured
strains in complete CIN medium. The present study was
carried out to determine the effect of individual and combined selective agents on the growth rates of plasmid-bearing and plasmid-cured Y. enterocolitica and to determine
the relationship between virulence plasmid expression,
incubation temperature and selective agents.
UK), 20 mg was dissolved in 20 ml of ethanol (Merck
Chemicals Ltd, Darmstadt, Germany).
Novobiocin (monosodium salt, Merck), 125 mg was dissolved in 20 ml of sterile distilled water.
Yersinia-selective supplement (CIN) (SR 109, Oxoid)
was prepared by the addition of 2 ml of a 1 : 1 ethanol and
sterile distilled water mixture to one vial of supplement.
Culture studies
MATERIALS AND METHODS
Yersinia enterocolitica culture
Y. enterocolitica GER, serotype O:3 P‡ and P± cultures
were supplied by Dr S. Bhaduri (USDA, ARS, ERRC,
Philadelphia, PA, USA). The P‡ cultures were stored frozen on beads (Protect, Technical Consultant Services Ltd,
Heywood, UK) and the P± cultures maintained on tryptone soya agar slopes and plates (TSA, CM 131, Oxoid,
Unipath Ltd, Basingstoke, UK) as described previously
(Sheridan et al. 1998).
Inoculum preparation and enumeration
The P‡ and P± inocula were prepared and enumerated as
described previously (Sheridan et al. 1998). Brie¯y, cultures of P‡ and P± strains were inoculated into 90 ml
volumes of brain heart infusion broth (BHI, CM 225,
Oxoid) and incubated overnight at 25 C. The cultures
were serially diluted in sterile, ®ltered maximum recovery
diluent (MRD, CM 733, Oxoid), and the numbers of Y.
enterocolitica per millilitre of suspension determined using a
membrane ®ltration epi¯uorescent staining technique
(Walls et al. 1989). The enumerated suspension was diluted
in MRD to give a working inocula and the cell concentrations were con®rmed on Yersinia-selective agar (CIN, CM
653, SR 109, Oxoid).
Broth preparation
Yersinia-selective broth base (CIN NA) was prepared from
®rst principles according to the Oxoid Manual (6th edition
1990) (Sheridan et al. 1998).
Selective supplements
Individual solutions of the selective agents of Yersiniaselective supplement were prepared on the day of use.
Cefsulodin (sodium salt, Sigma-Aldrich Co. Ltd, Poole,
UK), 75 mg was dissolved in 2 ml of sterile distilled water.
2, 2, 40 -trichloro-20 hydroxy diphenyl ether (Irgasan DP
300) (Ciba Geigy Dyestuffs and Chemicals, Manchester,
Two-hundred and twenty-®ve millilitre volumes of CIN
NA broth base were temperature equilibrated at 25 or 37 ‹ 1 C, in a thermostatically controlled water-bath (Lauda,
KoÈnigshofen, Germany) for approximately 1 h before use.
The broth was transferred to a sterile 250-ml nalgene ¯ask
(Nalgene, Nunc Int., Rochester, NY, USA) ®tted with a
sterile temperature probe and catheter needle (Becton and
Dickinson, Dublin, Ireland) attached. Aliquots (09 ml) of
the individual selective agents (i.e. cefsulodin, irgasan,
novobiocin), or combinations of these agents, or the commercial CIN supplement (SR 109) were mixed into the
broth base.
Pure culture systems were prepared by adding a 100-ml
inocula of Y. enterocolitica P‡ or P± strains containing
approximately 23 500 cfu mlÿ1 to the supplemented CIN
NA broths to give initial concentrations of approximately
1000 cfu mlÿ1. The mixtures were shaken thoroughly to
disperse the inocula within the broths. All culture systems
were incubated in water baths at 25 or 37 ‹ 1 C for 9 h.
Every hour, a 50-ml sample of culture was withdrawn
from each culture using a sterile syringe (Becton and
Dickinson) and examined as described below. Each experiment was repeated on three occasions.
Growth studies were carried out for the two Y. enterocolitica strains (P‡ and P±) in CIN NA broth, in CIN NA
broth supplemented with single selective agents (i.e. cefsulodin, irgasan, novobiocin), and with combinations of two
selective agents, as well as in CIN NA broth containing
CIN supplement.
Generation of growth curves
Volumes (10 ml) of the culture were serially diluted in
MRD and duplicate aliquots of each dilution series were
plated out on Yersinia-selective agar (CIN) using a spiral
plater (Model D, Spiral Plating Systems, Cincinnati, OH,
USA). Plates were incubated at 37 C for 24 h and the
numbers of Y. enterocolitica (cfu mlÿ1) colonies on each
plate were enumerated using image analysis (Seescan
Imaging Systems, Cambridge, UK). Counts obtained were
plotted against time to produce growth curves.
= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 88, 1001ÿ1008
GROWTH KINETICS OF YERSINIA
Statistical analysis
All growth rate data from these studies was analysed using
linear regression analysis (Systat, Evanston, IL, USA). The
lag phase durations and growth rates of each Y. enterocolitica strain were determined for each temperature/selective
agent regime according to Broughall et al. (1983). Brie¯y,
all data were included from the points at which the cell
concentration had increased to 150% of the inoculated concentration, to the point where the population density
ceased to increase. The speci®c lag times and growth rates
of Yersinia in the presence of the different supplements
were determined as previously described (Duffy et al.
1994).
Statistical analysis was carried out from the design of the
experiment which consisted of randomized blocks of data
containing two strains (P‡ and P±), eight enrichment
broths (CIN NA, CIN NA ‡ cefsulodin, CIN NA ‡ irgasan, CIN NA ‡ novobiocin, CIN NA ‡ cefsulodin and
irgasan, CIN NA ‡ cefsulodin and novobiocin, CIN NA ‡ irgasan and novobiocin, CIN NA ‡ CIN supplement), and
two incubation temperatures (25, 37 C) in factorial combination. This design was used for both lag phase and growth
rate data analysis. Analysis of variance (ANOVA) was carried out on lag phase and growth rate data using Genstat 5
(Rothamsted Experimental Station, Harpenden, UK). The
standard error of difference was determined from the
ANOVA. Results presented are a mean of three replicates.
t-tests were carried out to compare the lag phase durations
of the strains in different supplemented media, a similar
procedure was used for growth rate data. The signi®cance
of the differences between the means was determined using
the least signi®cance approach (Steel and Torrie 1980).
RESULTS
Table 1 shows the effect of single or combinations of selective agents in CIN NA broth on the lag phase duration of
Y. enterocolitica P‡ and P± strains incubated at 37 and 25 C. In general, lag phases were longer at 37 than 25 C. An
exception to this pattern was noted for Yersinia cultures in
CIN NA broth at 25 C where the lag phase could not be
determined (Sheridan et al. 1998). It was also noted that
the lag phases were generally longer in the presence of the
selective agents for both temperatures examined.
When the in¯uence of selective agents on the lag phase
durations of P‡ and P± strains were compared at 37 C;
signi®cantly longer lag phases (P < 001) were noted for
P‡ strains than P± strains in CIN NA broth supplemented
with cefsulodin and novobiocin. While, P± strains had a
signi®cantly longer (P < 005) lag phase in the presence of
irgasan or full CIN supplement.
1003
Table 1 The effect of selective agents on the lag phase durations
(h) of Yersinia enterocolitica
Incubation temperature ( C)
Broth
CIN NA
CIN NA ‡ cefsulodin
CIN NA ‡ irgasan
CIN NA ‡ novobiocin
CIN NA ‡ Cef ‡ Irg
CIN NA ‡ Cef ‡ Nov
CIN NA ‡ Irg ‡ Nov
CIN NA ‡ CIN supplement
37
P‡
P ±
25
P‡
P ±
041
045
104
096
081
134
175
088
051
057
165
085
078
064
140
158
NC
094
047
070
055
012
039
092
NC
069
109
062
074
028
050
069
NC ± not calculated, SED ± 024, CIN NA ± Yersinia-selective
broth base, degrees of freedom ˆ 58.
At 25 C, the lag phase duration of P‡ and P± strains
were generally similar. It was noted, however, that the lag
phase of P± strains was signi®cantly longer (P < 005) than
P‡ where irgasan was present.
The overall duration of the lag phase of Y. enterocolitica
P‡ and P± strains was in¯uenced by both the selective
agents present and the incubation temperature. Longest lag
phases for P‡ strains were noted in CIN NA broth supplemented with irgasan and novobiocin at 37 C, and for P±
strains in the presence of irgasan. At 25 C, P± strains had
the longest phase in the presence of cefsulodin or CIN supplement, while P± strains had the longest duration where
irgasan was present.
When the lag phase durations were compared between
the incubation temperatures, they were generally longer in
cultures incubated at 37 C. P‡ strains had signi®cantly
longer (P < 005) lag phase duration at 37 C where irgasan, or combinations of selective agents, i.e. cefsulodin and
novobiocin, irgasan and novobiocin, were present. In contrast, the lag phase duration for P‡ strains was signi®cantly
longer (P < 0 05) at 25 than 37 C in the presence of cefsulodin.
With P± strains, the lag phases were signi®cantly longer
at 37 than 25 C in CIN NA broth containing irgasan (P < 005) and with combinations of irgasan and novobiocin
and full CIN supplement (P < 0 001).
Examination of the growth rate data revealed a linear
relationship between bacterial numbers and time during
the growth phase of Y. enterocolitica P‡ and P± cells. High
correlation coef®cients (r2 values) were observed under
most growth conditions with r2 values ranging from 071 to
= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 88, 1001ÿ1008
1004 C . M . L O G U E E T A L .
Table 2 The effect of selective agents on the growth rates (cfu
mlÿ1 hÿ1) of Y. enterocolitica
Incubation temperature ( C)
Broth type
CIN NA
CIN NA ‡ cefsulodin
CIN NA ‡ irgasan
CIN NA ‡ novobiocin
CIN NA ‡ Cef ‡ Irg
CIN NA ‡ Cef ‡ Nov
CIN NA ‡ Irg ‡ Nov
CIN NA ‡ CIN supplement
37
P‡
P ±
25
P‡
P ±
033
029
038
041
014
025
012
040
037
043
038
040
028
034
023
028
016
034
034
032
031
032
034
029
012
034
032
034
033
028
030
028
Fig. 2 The growth of Yersinia in CIN NA broth ‡ CIN
supplement
SED ˆ 003, degrees of freedom ˆ 63.
099. The low r2 value (r2 ˆ 071), associated with the
growth of P‡ strains in CIN NA broth supplemented with
cefsulodin and irgasan, was attributed to a high RSD value,
indicating a large variation between replicates.
Table 2 shows the effect of selective agents on the
growth rate of Y. enterocolitica P‡ and P± strains.
Generally, the growth rates of Yersinia were faster at 37
than 25 C (Fig. 1). When the growth rates of P‡ and P±
strains were compared at 37 C, the plasmid-cured strain
(P±) grew signi®cantly faster (P < 001) than P‡ strain in
most media. Exceptions were noted in CIN NA ‡ CIN
supplement where the opposite case occurred with signi®cantly faster (P < 0001) growth rates noted for P‡ strains
(Fig. 2) and in the presence of irgasan and novobiocin
where no differences were observed. At 25 C, no differ-
Fig. 1 The growth of Yersinia in CIN NA broth ‡ novobiocin
ences were observed in the growth rates of P‡ or P±
strains.
The overall growth rate of the Yersinia strains was in¯uenced by both the selective agents present and the incubation temperature. P‡ strains grew fastest in CIN NA broth
at 37 C supplemented with novobiocin or CIN supplement, while P± strains grew fastest in the presence of single
selective agents, i.e. cefsulodin, irgasan or novobiocin. At
25 C, the growth rates of P‡ and P± strains were generally similar although slower growth rates were noted in
CIN NA broth for both strains.
When the growth rates of P‡ and P± strains were compared between temperatures, P‡ strains grew signi®cantly
faster (P < 001) at 37 than 25 C in CIN NA broth and in
the presence of novobiocin and full CIN supplement.
Signi®cantly faster growth rates were noted at 25 than at
37 C where combinations of supplements, i.e. cefsulodin
and irgasan, irgasan and novobiocin, and cefsulodin and
novobiocin (P < 005) were present.
With the plasmid-cured strain (P±) growth was signi®cantly faster (P < 0 05) at 37 C in the presence of most
supplements, except in the presence of irgasan and novobiocin where growth was faster at 25 C (P < 005) or with
full CIN supplement where no differences were observed.
Overall, selective agents extended the lag phase duration
of Y. enterocolitica P‡ and P± strains at 37 C, when one
or more combinations of selective supplements were present. When the growth rate data were examined, there was
an association between strain type and incubation temperature. Plasmid-cured strains generally grew faster than plasmid-bearing strains at 37 C in most media indicating
virulence plasmid expression at the higher temperature
retarded P‡ strains. One exception was observed, however,
where P‡ strains grew signi®cantly faster than P± in the
= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 88, 1001ÿ1008
GROWTH KINETICS OF YERSINIA
presence of CIN supplement, indicating a possible association between selective agents, incubation temperature and
virulence plasmid expression. In contrast, the growth rates
of the plasmid-cured strain appeared to be retarded at 37 C to rates similar to those observed at 25 C.
DISCUSSION
This study aimed to determine the individual and combined effects of selective agents on the growth kinetics of
plasmid-bearing (P‡) and plasmid-cured (P±) Y. enterocolitica cells and to elucidate the effects of strain growth
kinetics, strain type, selective agent(s) and incubation temperature.
Sheridan et al. (1998) reported that at 37 C, the lag
phase of Y. enterocolitica in CIN NA broth plus selective
agents was longer than the lag phase observed in CIN NA
broth. A similar pattern was observed in a study of Listeria
monocytogenes (Sheridan et al. 1994), which also reported
an extension of the lag phase duration in the presence of a
selective supplement. The current study con®rmed these
observations in relation to the effects at 37 C, however, no
extended lag phase duration was observed in supplemented
CIN NA broth at 25 C. The shortest lag phase occurred
at 25 C in the presence of cefsulodin and novobiocin.
Virulence plasmid expression did not appear to in¯uence
the lag phase duration. It is dif®cult to interpret these
effects, but temperature appears to play an important part.
The presence or absence of a virulence plasmid had an
effect on growth rate, i.e. plasmid-bearing Y. enterocolitica
strains grew slower than plasmid-cured cells at 37 C. This
observation is in agreement with previously reported
results (Sheridan et al. 1998), and may be associated with
the additional metabolic burden involved in the expression
of the virulence plasmid (McDermott et al. 1993; Goverde
et al. 1994).
This study observed that growth was slow in CIN NA
broth at 25 C. This may be due to the inhibitory nature of
the broth base, as previously reported (Fukushima and
Gomyoda 1986; Sheridan et al. 1998). A previous study by
Sheridan et al. (1998) showed that the inhibition could be
reduced by increasing the incubation temperature to 37 C
or by inclusion of the two antibiotics and the surfactant,
irgasan, in the medium at 25 C. The present study con®rmed these observations, and also found that the inhibition could be reduced by the addition of one or more of
the selective agents (cefsulodin, irgasan or novobiocin).
Sheridan et al. (1998) suggested a number of possible
reasons for the absence of inhibition under such conditions.
These included (a) the production of substances by the
cells that complex the inhibitory agent or limit its activity
or (b) as a result of changes in cell structure that prevent
or reduce the access of inhibitors into the cell. Both these
1005
processes could be operational in cultures grown at 37 C.
When plasmid-bearing Yersinia cells are grown at 37 C,
they produce signi®cant amounts of Yersinia, outer membrane proteins (YOPs) which could bind the inhibitor
(Bhaduri 1990; Cornelis 1992). It is well known that most
cells adapt the composition and characteristics of the cell
membrane in response to environmental changes, speci®cally producing a more rigid cell membrane at 37 C than
at 25 C, these changes can result in reductions in membrane permeability (Tsuchiya et al. 1987; Nagamachi et al.
1991). This effect is associated with the levels of saturated
and unsaturated fatty acids present in the outer membranes. At higher temperatures, the saturated fatty acid
content increases and unsaturated fatty acids decrease
(Nagamachi et al. 1991). Membranes with increased saturated fatty acids are less permeable than membranes with
high unsaturated fatty acid contents (Cornelis et al. 1987;
Tsuchiya et al. 1987; Nagamachi et al. 1991; Goverde et al.
1994; Bodnaruk and Golden 1996). As well as these
changes in lipid content, incubation at higher temperatures
causes a number of qualitative and quantitative changes in
membrane-associated proteins. Outer membrane porins are
no longer synthesized (Cornelis et al. 1987; Rhode et al.
1994). These porins have an important role in the transport
of a range of substances, including antibiotics across the
outer cell membrane (Brown and Williams 1985; Brozostek
and Hrebenda 1988). Thus, at the higher temperatures,
reductions in porin concentrations would reduce the access
of selective agents to the cell interior, limiting the inhibitory impact of such agents on cellular growth.
The calcium content of the growth medium may also
play a role in modulating the effects observed in this study.
At 37 C, plasmid-bearing (P‡) Y. enterocolitica cells
require speci®c concentrations of Ca2‡ in growth media to
support normal growth. In the absence of adequate calcium, cells can cease growing and produce a number of
YOPs which are released into the growth medium. YOPs
are encoded by the virulence plasmid and their expression
is temperature and calcium regulated (Cornelis et al. 1987;
Bhaduri 1990; Michiels et al. 1990; Li et al. 1998).
Sheridan et al. (1998) reported that the calcium level in
CIN NA medium was 325 mmol lÿ1 Li et al. (1998) recently
reported that Ca2‡ levels above 745 mmol lÿ1 were required
to preserve virulence plasmid and prevent YOP production, it is therefore likely that the calcium concentration in
the CIN NA broth would induce a low calcium response.
At 25 C, two major effects were observed for plasmidcured (P±) strains, similar to cultures incubated at 37 C,
i.e. growth was not prevented by any of the antibacterial
agents alone, or in combination with irgasan, growth inhibition in the CIN NA medium did not occur. As noted
previously, cells incubated at 25 C increase their permeability by altering the outer membrane. This is facilitated
= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 88, 1001ÿ1008
1006 C . M . L O G U E E T A L .
by changes in the levels of unsaturated fatty acids, and an
increased synthesis of porins. Despite this, neither of the
antibiotics or irgasan used in the present study resulted in
inhibition of cell. This is in line with previous reports on
the effects of novobiocin (Raevuori et al. 1978; Schiemann
1980; Toora et al. 1994), cefsulodin (Toora et al. 1994) a
combination of irgasan and novobiocin (Toora et al. 1994)
or all three substances together (Toora et al. 1994). In their
work at 25 C Toora et al. (1994) showed that all three substances added to the medium together were more inhibitory
than single or double antibiotics, which was generally the
case in the present study.
Given the increased membrane permeability of cells and
increased porin carrier availability at 25 C, it would be reasonable to assume that the selective agents should have
greater impact in reducing growth rates at this temperature, however, this was not the case. The general lack of
inhibition by antimicrobials and the negation of inhibition
observed in unsupplemented media by the creation of a
more inhibitory environment are dif®cult to explain. They
may be the net outcome of a number of conditions that
apply to Yersinia cells incubated at 25 C. While porins are
present at this temperature, substances such as cefsulodin,
novobiocin and irgasan are hydrophobic which results in
their having a low cell permeability (Nikaido and Nakae
1979; Meincke et al. 1980; Nikaido et al. 1983).
The O-antigen glycoside side chains of Yersinia are
longer and project further from the surface of the organism
in strains cultured at 25 C, compared with similar strains
incubated at 37 C (Skurnik and Toivanen 1993; Goverde
et al. 1994). These extended chains may present a more
signi®cant steric barrier limiting the entry of substances
into the cell, alternatively, such external chains may interact with selective agents preventing them from continuing
through the cell membrane to their site of action within the
cell (Skurnik and Toivanen 1993; Goverde et al. 1994).
Steric protection of Yersinia enterocolitica cells by cell surface components has been reported by Martinez (1983)
who noted that cells cultured at 25 C were able to withstand the killing action of serum compliment, compared
with similar cells cultured at 37 C. Such cell surface/capsule adaptations could limit or prevent the entry of inhibitory substances including antibiotics and surfactants into
Yersinia cells.
At 37 C, the addition of single antibiotics generally had
no effect on growth rate of Y. enterocolitica. Overall growth
at 37 C was faster than at 25 C, indicating an increased
susceptibility of Yersinia to antibiotics at lower temperatures (Kouwatli et al. 1979). In keeping with the data at 25 C, virulence plasmid expression did not have any effect on
growth rates, except in the presence of cefsulodin, where
plasmid expression signi®cantly decreased the growth rate.
According to Robins-Browne et al. (1986) in the presence
of the virulence plasmid the minimum inhibitory concentrations of a number of antibiotics to Yersinia enterocolitica,
including novobiocin, was signi®cantly less at 37 C in plasmid-bearing cells. A similar result has also been reported
by Jimenez-Valera et al. (1989) for other antibiotics. In studies of E. coli it was shown that the presence of a Col V
plasmid increased the sensitivity of the organism to novobiocin and rifampicin (Davies et al. 1986).
In this study the most notable effects observed were
when each of the antibiotics and the surfactant were combined together in the growth medium at 37 C. In the presence of the virulence plasmid, all combinations were
signi®cantly more inhibitory on growth and the data show
that irgasan greatly enhanced the inhibitory effect of both
antibiotics.
It has been shown that drugs may act synergistically
with surfactants to increase their activity on cells. This
may result in an increased transfer of the drug across the
outer membrane or in an increase in the binding of the
drug to the receptor sites (Attwood and Florence 1985).
Irgasan acts as a site-directed inhibitor of enoyl-acyl carrier
protein reductase by mimicking its natural substrate, thus
inhibiting fatty acid synthesis and, consequently, lipid formation (Levy et al. 1999). Such inhibition will dramatically
affect the formation of the outer membrane and increase
antibiotic transfer across the barrier. As well as these
effects, surfactants may also bind proteins in solution and
the possible signi®cance of these interactions could explain
some the inhibitions observed in this study.
Although the synergistic activity between irgasan and
the antibiotics is a possible method of activity of surfactants
and antibiotics inhibiting cells, it should be noted that this
type of inhibition was only observed for plasmid-bearing
cells. This would support the view that some form of protein/surfactant interaction resulted in an increased antibiotic activity. A possible mechanism for this may be as
follows.
Irgasan is hydrophobic and lipophilic in nature, and suf®ciently soluble to be absorbed into the hydrophobic areas
of cell walls and membranes. According to Meincke et al.
(1980) irgasan is adsorbed to the outer cell membrane
which behaves as a barrier to its entry into the cytoplasmic
membrane, which is its site of action. It has been suggested
(Meincke et al. 1980) that irgasan is adsorbed onto the
outer cell membrane which prevents penetration to and
damage of the inner cytoplasmic membrane. Such irgasan/
membrane complexes could interact with large amounts of
outer membrane proteins present in outer membranes of
plasmid-bearing Yersinia growing at 37 C, disrupting protein membrane structures, leading to increased membrane
permeability (Attwood and Florence 1985). Cell membranes damaged in this way could be more easily penetrated by the antibiotics, which would mean that such
= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 88, 1001ÿ1008
GROWTH KINETICS OF YERSINIA
antibiotics would more easily gain access to and bring
about damage of the cytoplasmic membrane. Alternatively,
the important interactions between irgasan and antibiotics
may occur after they have separately penetrated the cell.
Irgasan as the surfactant may increase the avidity of the
antibiotics to their sites of action within the cell (Attwood
and Florence 1985).
This study noted that under some circumstances the
presence of antibiotic increased the growth rates of cells.
Thus, growth rates in the presence of each antibiotic are
among the highest noted within the study. As antibiotics
are normally viewed as suppressing bacterial growth, these
observations would appear, on ®rst analysis, to be unusual.
However, there are previous reports of the enhancement of
growth in the presence of antibiotics. Novobiocin has been
previously reported to cause increases in the growth rates
of cells (Drilica and Rouviere-Yaniv 1993) and this has
been associated with the enhancement of enzyme/DNA
gyrase activities in growing cells (Drilica and RouviereYaniv 1993; Sheridan et al. 1998).
Rapid growth was also observed in other circumstances
where inhibition may have been predicted. Thus, in complete CIN, i.e. CIN NA broth with full selective supplement, the presence of both antibiotics and the surfactant,
might have been expected to be considerable if not maximal. This phenomenon could not be explained satisfactorily
but emphasizes that complex interactions occur between
antibiotics and surfactants and the major in¯uence that the
virulence plasmid expression can have on the growth of the
organism is at 37 C.
This study raises questions in relation to the development of media for the growth or isolation of pathogens
from foods. Despite recent advances, the ability to rapidly
and comprehensively recover and/or differentiate organisms in liquid or agar media still remains the cornerstone
of many areas of microbiology (Bloom®eld et al. 1998).
The study raises questions in relation to the development
of media for the growth or isolation of Y. enterocolitica or
other signi®cant pathogens from foods. Depending on the
net effect of the interactions of physiological and genetic
status of the strains, the synergistic and antagonistic interactions between media components and other aspects of
cultural conditions growth may be signi®cantly enhanced
or reduced. Of greater concern are the observations in this
study which demonstrate that the construction of media is
not simply a process of identifying contaminants that need
to be suppressed and the selection of agents which are
more active against the contaminant than the target organism, and combining these agents in a cost-effective format.
More important is examining the effect of adding substances together in media without determining their combined effects. The results of selecting multiple antibiotics
may be inhibitory, while if present alone they will have no
1007
effect. In this study it was shown that the presence of surfactants may increase or decrease the effects of other inhibitory agents, in unpredictable ways. Such effects are
rarely taken into account on the development or application
of selective media. Yet the availability of media for the
rapid, ef®cient selection and cultivation of bacterial species
of interest remains central to most microbiological research
and practice. In conclusion, further detailed research into
the phenotypic and genetic status of bacteria and their physiological interactions with current and potential components of selective media is important in the development of
more ef®cient methods for the recovery, isolation and analysis of many important bacterial species.
ACKNOWLEDGEMENT
The authors gratefully acknowledge the ®nancial assistance
of the European Union.
REFERENCES
Attwood, D. and Florence, A.T. (1985) Surfactant Systems, Their
Chemistry, Pharmacy, and Biology. p. 447. London: Chapman &
Hall.
Bhaduri, S. (1990) Calcium-response expression of plasmidmediated outer membrane proteins from Yersinia enterocolitica
grown on solid media. Journal of Industrial Microbiology 5, 207±
214.
Bhaduri, S., Turner-Jones, C., Taylor, M.M. and Lachica, R.V.
(1990) Simple assay of calcium dependency for virulent plasmid-bearing clones of Yersinia enterocolitica. Journal of Clinical
Microbiology 28, 798±800.
Bloom®eld, S.F., Stewart, G.S.A.B., Dodd, C., .E.R., Booth, I.R.
and Power, E.G.M. (1998) The viable but non-culturable phenomenon explained? Microbiology 144, 1±3.
Bodnaruk, P.W. and Golden, D.A. (1996) In¯uence of pH and
incubation temperature on fatty acid composition and virulence
factors of Yersinia enterocolitica. Food Microbiology 13, 17±22.
Broughall, J.M., Anslow, P.A. and Kilsby, D.C. (1983) Hazard
analysis applied to microbial growth in foods: Development of
mathematical models describing the effect of water activity.
Journal of Applied Bacteriology 55, 101±110.
Brown, M.R.W. and Williams, P. (1985) In¯uence of substrate
limitation and growth phase on sensitivity to antimicrobial
agents. Journal of Antimicrobial Agents and Chemotherapy 15
Suppl. A, 7±14.
Brozostek, K. and Hrebenda, J. (1988) Outer-membrane permeability to b-lactam antibiotics in Yersinia enterocolitica. Journal
of General Microbiology 134, 1535±1540.
Cornelis, G.R. (1992) Yersiniae, ®nely tuned pathogens. In
Molecular Biology of Bacterial Infection: Current Status and
Future Perspectives ed. Hornaeche, C.E., Penn, C.W. and
Smyth, C.J. p. 231. Cambridge: Cambridge University Press.
= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 88, 1001ÿ1008
1008 C . M . L O G U E E T A L .
Cornelis, G., Laroche, Y., Balligand, G., Sory, M.-P. and
Wauters, G. (1987) Yersinia enterocolitica, a primary model for
bacterial invasiveness. Reviews of Infectious Diseases 9, 64±87.
Davies, C., Somanth, J., Sushela, D. and Rowbay, R.J. (1986)
Virulence plasmid-associated sensitivity to rifampicin and novobiocin in Escherichia coli. Letters in Applied Microbiology 2, 119±
123.
Drilica, K. and Rouviere-Yaniv, J. (1993) Histone-like proteins of
bacteria. Microbiological Reviews 51, 301±319.
Duffy, G., Sheridan, J.J., Buchanan, R.L., McDowell, D.A. and
Blair, I.S. (1994) The effect of aeration, initial inoculum and
meat micro¯ora on the growth kinetics of Listeria monocytogenes
in selective enrichment broths. Food Microbiology 11, 429±438.
Fukushima, H. and Gomyoda, M. (1986) Growth of Yersinia
pseudotuberculosis and Yersinia enterocolitica biotype 3B serotype
O:3 inhibited on cefsulodin-irgasan-novobiocin agar. Journal of
Clinical Microbiology 24, 116±120.
Goverde Rosalina, L.J., Kusters, J.G. and Huis in't Veld, J.H.J.
(1994) Growth rate and physiology of Yersinia enterocolitica;
in¯uence of temperature and presence of the virulence plasmid.
Journal of Applied Bacteriology 77, 96±104.
Jimenez-Valera, M., Ruiz-Bravo, A., Roman, S.M. and RaniosCormenza, A. (1989) Increase of the antimicrobial susceptibility
of Yersinia enterocolitica associated with the expression of the
virulence plasmid. Current Microbiology 18, 211±214.
Kouwatli, K., Bejar, V., Ruiz-Brao, A. and Ramos-Cormenzana,
A. (1979) Sensitivity of Yersinia enterocolitica to several antimicrobial agents. Microbios Letters 11, 137±142.
Laird, W.J. and Cavanaugh, D.C. (1980) Correlation of autoagglutination and virulence of Yersiniae. Journal of Clinical
Microbiology 11, 430±432.
Levy, C.W., Roujeinikova, A., Sedelnikva, S., Baker, P.J., Stuitje,
A.R., Slabas, A.R., Rice, D.W. and Rafferty, J.B. (1999)
Molecular basis of triclosan activity. Nature 398, 383±384.
Li, H., Bhaduri, S. and Magee, W.E. (1998) Maximising plasmid
stability and production of released proteins in Yersinia enterocolitica. Applied and Environmental Microbiology 64, 1812±1815.
Martinez, R.L. (1983) Plasmid-mediated and temperature-regulated surface properties of Yersinia enterocolitica. Infection and
Immunity 41, 921±930.
McDermott, P.J. and Gowland Pauline, Gowland, P.C. (1993)
Adaptation of Escherichia coli growth rates to the presence of
pBR322. Letters in Applied Microbiology 17, 139±143.
Meincke, B.E., Kranz, R.G. and Lynch, D.L. (1980) Effect of
irgasan on bacterial growth and its adsorption into the cell wall.
Microbios 28, 133±147.
Michiels, T., Wattiau, P., Brasseur, R., Ruysschaert, J.M. and
Cornelis, G. (1990) Secretion of Yop proteins by Yersiniae.
Infection and Immunity 58, 2840±2849.
Nagamachi, E., Shibuya, S.-I., Hirai, Y., Matsushita, O.,
Romochika, K.-I. and Kanemasa, Y. (1991) Adaptational
changes of fatty acid composition and the physical state of
membrane lipids following the change of growth temperature in
Yersinia enterocolitica. Microbiology and Immunology 35, 1085±
1093.
Nikaido, H., Rosenberg, E.Y. and Foulds, J. (1983) Porin
Channels in Escherichia coli: Studies with b-Lactam intact cells.
Journal of Bacteriology 153, 232±240.
Nikaido, H. and Nakae, T. (1979) The outer membrane of gramnegative bacteria. Advances in Microbial Physiology 4, 163±250.
Raevuori, M., Harvey, S.M., Pickett, M.J. and Martin, W.J.
(1978) Yersinia enterocolitica: In vitro antimicrobial susceptibility. Antimicrobial Agents and Chemotherapy 15 Suppl A. , 7±14.
Rhode, J.R., Fox, J.M. and Minnich, S.A. (1994)
Thermoregulation in Yersinia enterocolitica is coincident with
changes in DNA supercoiling. Molecular Microbiology 12, 187±
199.
Robins-Browne, R.M., Prpic, J.K. and Davey, R.B. (1986)
In¯uence of the virulence plasmid and the congo red reaction
on the antimicrobial susceptibility of Yersinia species. Journal of
Antimicrobial Chemotherapy 17, 553±557.
Schiemann, D.A. (1979) Synthesis of a selective agar medium for
Yersinia enterocolitica. Canadian Journal of Microbiology 25,
1298±1304.
Schiemann, D.A. (1980) Yersinia enterocolitica: observations on
some growth characteristics and response to selective agents.
Canadian Journal of Microbiology 26, 1232±1240.
Schiemann, D.A. (1982) Isolation of Yersinia enterocolitica. Oxoid
Culture 3, 1.
Schiemann, D.A. (1989) Yersinia enterocolitica and Yersinia pseudotuberculosis. In Foodborne Bacterial Pathogens. ed. Doyle, M.P.
pp. 601±672. New York, NY: Marcel Dekker, Inc.
Sheridan, J.J., Logue, Catherine, M., McDowell, D.A., Blair, I.S.
and Hegarty, T. (1998) A study of the growth kinetics of
Yersinia enterocolitica serotype O:3 in pure and meat culture
systems. Journal of Applied Microbiology 85, 293±301.
Sheridan, J.J., Duffy, G., Buchanan, R.L., McDowell, D.A. and
Blair, I.S. (1994) The use of selective and non-selective enrichment broths for the isolation of Listeria species from meat. Food
Microbiology 11, 439±446.
Skurnik, M. and Toivanen, P. (1993) Yersinia enterocolitica lipopolysaccharide: genetics and Virulence. Trends in Microbiology
1, 148±152.
Steel, R.G.D. and Torrie, J.H. (1980) Principles and Procedures of
Statistics. New York, NY: McGraw-Hill.
Toora, S., Budu-Amoako, E., Ablett, R.F. and Smith, J. (1994)
Evaluation of different antimicrobial agents used as selective
supplements for the isolation and recovery of Yersinia enterocolitica. Journal of Applied Bacteriology 77, 67±72.
Tsuchiya, H., Sato, M., Kanematsu, N., Kato, M., Hoshino, Y.,
Takagi, N. and Namikawa, I. (1987) Temperature-dependent
changes in phospholipid and fatty acid composition and membrane lipid ¯uidity of Yersinia enterocolitica. Letters in Applied
Microbiology 5, 15±18.
Walls, I., Sheridan, J.J., Welch, R.W. and McDowell, D.A.
(1989) A rapid method of enumerating microorganisms from
beef using an acridine orange direct count technique. Irish
Journal of Food Science and Technology 13, 23±31.
= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 88, 1001ÿ1008